Video Cards & GPUs News - Page 65

NVIDIA's RTX 4060 and its Ti version are supposedly nearing launch, as well as AMD's RX 7600 graphics card, and a YouTube leaker has been discussing what he's heard regarding the potential pricing for these GPUs.

Of course, if the rumor mill is right, as we've just seen courtesy of VideoCardz, the theory is that the RTX 4060 Ti and the RX 7600 will be launching almost simultaneously - a day apart, in a couple of weeks' time - with the RTX 4060 following in two months (early July).

Price-wise, though, according to RedGamingTech (RGT) on YouTube, it's the RTX 4060 that'll be pitched at about the same price as AMD's RX 7600, with the RTX 4060 Ti getting a price tag of around $100 more expensive in the US.

NVIDIA's Lovelace flagship GPU has been spotted with a massive overclock applied, and not for the first time - however, it is the first time that such a significant ramping up of boost speed has been seen while running a 3D application.

The RTX 4090 in question was a Colorful model, its iGame LAB flagship card, which was overclocked to 3.825GHz by expert overclocker 'CENS' using liquid nitrogen cooling.

The RTX 4090 was run at that speed in the Unigine Superposition benchmark, which may not be a gaming test, but it is a full 3D test that pushes the GPU as a result. When running with such high clocks, that's obviously going to add quite some strain. What's more, the benchmark was undertaken at 8K resolution to make for a steep hill to climb.

NVIDIA's RTX 4060 Ti graphics card could turn up earlier than expected, with the rumor mill now floating a release date that might be designed to beat AMD to the mid-range GPU punch.

Previously - and actually for a while now - we've been hearing that NVIDIA's RTX 4060 Ti will be launched at Computex, which is at the very end of May.

However, the latest chatter from the GPU grapevine insists that this date has been moved forward for the launch of the first version of the RTX 4060 Ti (which will have 8GB of VRAM).

A new image over at Videocardz.com showcases the new AMD Radeon RX 7600 graphics cards in stock at an undisclosed retail store in Asia. The Sapphire Pulse model confirms that AMD's mainstream RDNA 3 offering will feature 8GB of VRAM alongside 32 CUs (or Compute Units) with a fully unlocked Navi 33 GPU with 2048 Stream Processors.

This single image confirms that AMD is prepping to launch the Radeon RX 7600 later this month, where it will no doubt be compared to NVIDIA's upcoming GeForce RTX 4060 Ti - which is also set to debut with 8GB of VRAM. The only thing missing is an image of the GPU itself, with the packaging not showcasing the physical design and cooling.

Recent rumors for NVIDIA's GeForce RTX 4060 Ti are pointing to a version coming in July that will feature 16GB of VRAM. This additional model could be a direct response from NVIDIA to the current debate surrounding VRAM capacity amongst the PC gaming community. Although that info comes from a single source, it will be interesting to see if AMD follows suit for the Radeon RX 7600.

NVIDIA's RTX 4070 graphics card may soon be produced with a different GPU, or that's the latest we've heard on the grapevine.

The rumor is from Twitter and well-known leaker Kopite7kimi, who tells us in a short and sweet tweet that the RTX 4070 could have a variant coming based on the AD103 chip.

That's the GPU engine that drives the RTX 4080 (in slightly cut-down form from its full incarnation), and soon, the chip could be used in the RTX 4070 - but for what reason?

Although it's not official, info on the next GeForce RTX 40 Series game bundle has leaked via US retailers Ironside Computers and B&H Photo Video Audio. And it's a big one with one of the most anticipated PC releases in years, Diablo 4, set to be given away with the purchase of a GeForce RTX 4090, 4080, 4070 Ti, or 4070 desktop or GPU.

As a leak, the full terms and conditions have yet to be made available, but this is excellent news for those currently looking at upgrading in time to revisit Sanctuary when the game launches in June. We know this is set to begin this week, with the promotion starting May 9 and ending June 13, 2023.

Even better is confirmation that in addition to a game code for Diablo 4 valued at USD 69.99, newly minted GeForce RTX 40 Series owners will also get access to preorder bonus cosmetics for the game and many other Blizzard titles.

Rumors and leaks have been pointing to a GeForce RTX 4060 Ti launch later this month for quite some time now, with specs pointing to the GPU launching with 8GB of GDDR6 video memory on a 128-bit memory bus. Other specs that we previously reported on for the new GPU include info that it will make use of the AD106-350 chip with 4352 CUDA Cores.

With it being the most affordable entry-point in the GeForce RTX 40 Series line-up to date, a mainstream option depending on the MSRP, the main concern among the PC gaming community is the adoption of 8GB of VRAM at a time when new AAA PC releases are requiring more and more memory to enable high-end visual effects. The general consensus is that a high-powered GPU in 2023 should feature more than 8GB of VRAM.

Well, it looks like NVIDIA has been listening. A new report states that in addition to the GeForce RTX 4060 Ti 8GB launch happening later this month, NVIDIA has plans to release a 16GB model in July, with this variant to be announced alongside the GeForce RTX 4060 Ti's reveal in late May.

Here's a thing that we probably all do when it comes to testing out a brand-new GPU - briefly touch the backplate to see how hot it is to the touch. And here's the thing, GPU backplates are designed to get hot; they cover the back of the PCB, and for larger GPUs, they offer heat dissipation and structural support.

Still, you can see the logic behind these strange backplate-cooling solutions sold on AliExpress. If it's PC hardware-related and gets hot, try to cool it down with a few fans. And so, overclocker and tech reviewer der8auer has tested some of these new backplate coolers to see what they do.

As you can probably guess from the headline, the end result is little more than adding extra weight and physical size to a GPU. With AliExpress being home to several tech brands we've never heard of, der8auer begins the testing by taking a closer look at one of the coolers, and immediately the physical makeup is a little cheap. Sharp edges, glue for the fan mounting, zip ties, and adhesive tape for attaching it to a GIGABYTE Radeon RX 6800 XT.

Picking up a new GPU on the second-hand market is not uncommon, even for a product as fresh as AMD's Radeon RX 7900 XTX. That said, this is a strange one, as someone on Reddit posted an image of a second-hand Radeon RX 7900 XTX they purchased with an 'Internal Testing Only' label.

"Is there anything I should be aware of?" Redditor 'clumsyhotdog' wrote in the post. Many immediately pointed out that this could be an early version of the GPU that could feature some hardware differences from the retail model that shipped in December 2022. The user notes that it was purchased with a sizable discount compared to the MSRP of the GPU.

Pre-release models for GPUs are part of the process, with extensive internal testing to ensure drivers and other features are ready for launch. A pre-release Radeon RX 7900 XTX model could require special drivers and have features not present in the final retail release.

With strict export rules imposed by the United States government in 2021 prohibiting the export of US-based technologies (limiting the total performance of supercomputer hardware to China), we now have some specs and details for NVIDIA's mysterious A800 GPU developed for these constraints.

According to a new article at MyDrivers (via Tom's Hardware), the A800 operates at roughly 70% of the speed of NVIDIA's A100 GPU - which complies with the processing power limits in place for exports.

The NVIDIA A100 Tensor Core GPU is three years old, though still powerful in its own right, with 9.7 TFLOPS of FP64 and 19.5 TFLOPS of FP32 performance for the Ampere-based GPU. There's also the BFLOAT16 and FP16 Tensor Core performance of 624 TFLOPS with sparsity for AI workloads.